Ali Emre Gök

Ali Emre Gök

Auto-generated profile, based on publicly available information

Ali Emre Gök

Computer Engineering Department Hacı Bektaş Veli University Nevsehir Türkiye

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 July 2024 in Çukurova Üniversitesi Mühendislik Fakültesi Dergisi
Verified
Journal Article
Open access
PDF file
GBGülbahar Bilgiç
Gülbahar Bilgiç
AGAli Emre Gök
Ali Emre Gök
The need for energy in the world is increasing day by day and various energy production methods are used to meet this need. Production of hydrogen from biomass is one of these methods. Hydrogen production from biomass is a promising process to produce hydrogen and energy which has advantages such as the ability to use sustainable energy sources like biomass and solid waste, being carbon neutral, and increasing energy independence thanks to the variation of resources and the availability of local resources. The catalysts used in this process which can be conducted in three separate ways, affect hydrogen and energy production positively or negatively. One of the most important steps in effectively acquiring the ideal amount of product is predicting the outcomes of this procedure. This article compares a support vector regression (SVR) and random forest (RF) model to predict how various inputs used to produce hydrogen from biomass will affect hydrogen output. Additionally, the effect of catalyst addition on hydrogen yield in biomass processes was examined. In this context, 57 experimental studies from the literature were selected as a data set. From this data, 90% was selected for training and 10% for testing. The outputs were evaluated according to parameters such as R2, RMSE and MSE. The results show that RF and SVR models can significantly predict catalyst activity and hydrogen production.

13 July 2026 in Food Science & Nutrition
Verified
Journal Article
Open access
MYMustafa Yurdakul
Mustafa Yurdakul
ZBZeynep Sena Baştuğ
Zeynep Sena Baştuğ
AGAli Emre Gök
Ali Emre Gök
STSakir Tasdemir
Sakir Tasdemir
The strawberry (Fragaria × ananassa), known worldwide for its economic value and nutritional richness, is a widely cultivated fruit. Determining the correct ripeness level during the harvest period is crucial for both preventing losses for producers and ensuring consumers receive a quality product. However, traditional methods, that is, visual assessments alone, can be subjective and have a high margin of error. Therefore, computer‐assisted systems are needed. However, the scarcity of comprehensive datasets accessible to everyone in the literature makes it difficult to compare studies in this field. In this study, a new and publicly available strawberry ripeness dataset, consisting of 566 images and 1169 labeled objects across three ripeness classes (fully‐ripe, semi‐ripe, and unripe), prepared under variable light and environmental conditions in two different greenhouses in Turkey, is presented to the literature. Comparative tests conducted on the dataset using twelve YOLOv8, YOLOv9, and YOLO11 detection variants showed that the highest precision value was 90.94% in the YOLOv9c model, while the highest recall value was 83.74% in the YOLO11s model. In terms of the general performance criterion mAP@50, YOLOv8s was the best performing model with a success rate of 86.09% (mAP@50:95 = 75.52%). On this dataset, smaller and medium‐sized models (approximately 10–12 million parameters, e.g., YOLOv8s and YOLO11s) achieved accuracy comparable to the largest variants at substantially lower computational cost and higher inference speed, establishing a reproducible reference point for smart‐agriculture applications. These findings are specific to the present dataset and should be validated on larger, more diverse data before being generalized.

25 June 2026 in Research Square
Verified
Preprint
Open access
PDF file
AGAli Emre Gök
Ali Emre Gök
MYMustafa Yurdakul
Mustafa Yurdakul
STSakir Tasdemir
Sakir Tasdemir
Although Vision Transformer (ViT) architectures have resolved the global-context problem in medical image analysis, their quadratic computational complexity (O(N^2 )) creates hardware bottlenecks for high-resolution and 3D volumetric data. Mamba and Selective State-Space Models (Selective SSMs) provide an innovative solution to this problem by processing sequential data with linear computational complexity (O(N)). This systematic review was designed to examine the rapid adoption of the Mamba architecture in the field of medical artificial intelligence. In accordance with the PRISMA guidelines, 445 recent studies published between 2024 and 2026 were divided into four principal architectural typologies (Pure Mamba, CNN-Mamba Hybrids, Transformer-Mamba Integrations, and Novel Scanning Mechanisms) and passed through a Qualitative Critical Appraisal filter. The findings show that the most dominant and efficient approach in the literature is the CNN-Mamba hybrid, which combines local feature extraction with global context. Despite high diagnostic accuracy, the principal obstacles delaying clinical integration are the loss of spatial information arising from the flattening of 2D and 3D images into 1D sequences, the excessive memory consumption introduced by multi-directional scanning mechanisms, and the lack of multi-center external validation. Consequently, with the development of native 3D Mamba blocks that preserve spatial topology, the Mamba architecture is anticipated to become the backbone of next-generation foundation models in medical image analysis.